Wholly aromatic polyester resins have long been known. For instance, 4-hydroxybenzoic acid homopolymer and copolymers have been described in the past and are commercially available. Such polymers commonly are crystalline in nature and, when molten, frequently exhibit orientation in the melt; however, they have relatively high melting points or possess a decomposition temperature which is below the melting point, which leads to great difficulty in processing.
The homopolymer of p-hydroxybenzoic acid is a very high melting, insoluble material and, hence, very difficult to fabricate. Melting points as high as 610.degree. C. were quoted--see W. J. Jackson, The British Polymer Journal, December 1980, p. 155. In order to depress the high melting point of the homopolymer so as to make it melt fabricable, a variety of materials incorporating different types of comonomers were prepared over the years.
One such material is, for example, the resin made from p-hydroxybenzoic acid, isophthalic and/or terephthalic acids and 4,4'-biphenol as described in Cottis et al., U.S. Pat. Nos. 3,637,595 and 3,975,487. The polymer has outstanding high temperature properties and can be molded to give articles of high modulus and strength. It is offered commercially by Amoco Performance Products, Inc. under the trademark of Xydar.RTM..
The main drawback of the prior art p-hydroxybenzoic acid copolymers is the relatively high cost associated with the use of an expensive comonomer, such as 4,4'-biphenol, substituted hydroquinones (e.g., phenylhydroquinone), naphthalene diols, naphthalene dicarboxylic acids, and hydroxy-naphthoic acids. Efforts to replace these expensive monomers with the significantly less expensive hydroquinone, which is disclosed as an equivalent of biphenol in the aforementioned U.S. Pat. Nos. 3,637,595 and 3,975,487, were made by several research groups; however, none of these investigations were successful.
Study of the prior art shows that replacement of 4,4'-biphenol with hydroquinone leads to materials with inferior properties. The problem created by the introduction of hydroquinone is basically the following: at high terephthalate contents, high melting generally intractable polymers are obtained; tractability may be achieved at higher isophthalate levels, but the polyesters are relatively low melting and often display low second order glass transition temperatures which lead to low moduli and low heat distortion temperatures. For example, polyesters from p-hydroxybenzoic acid (PHBA) isophthalic acid (IA) and hydroquinone (HQ) were prepared by Deex, U.S. Pat. No. 4,377,681. At mole ratios PHBA/IA/HQ of 33.3/33.3/33.3 the material had a glass transition temperature of 110.degree. C.; when the above coreactants were used at ratios of 50/25/25, a Tg of 115.degree. C. was obtained.
The high melting points of a series of p-hydroxybenzoic acid/terephthalic acid/hydroquinone copolymers are graphically illustrated in FIG. 2 of the paper by G. W. Calundann, Industrial Development of Thermotropic Polyesters in High Performance Polymers: Their Origin and Development 235-249 (R. B. Seymour and G. S. Kirshenbaum, editors 1986). The publication shows clearly (in FIG. 2) that hydroquinone polymers melt at considerably higher temperatures than their 4,4'-biphenol counterparts. The T.sub.m of the lowest melting composition shown is about 420.degree. C. FIG. 4 of the same publication (p. 243) indicates how one research group was able to depress the melting points of the subject polymers by incorporating naphthalene diols, naphthalene dicarboxylic acids, and hydroxy naphthoic acids into them. From a purely technical point of view, the latter approach was a success; however, the modified polymers were still expensive due to the high cost of the naphthalene-based monomers.
The intractability of the hydroquinone-derived materials is discussed in Jackson et al., U.S. Pat. No. 4,242,496. Column 2, lines 18-26, states:
"U.S. Pat. No. 3,637,595 discloses that aromatic liquid crystal polyesters prepared from terephthalic acid, hydroquinone and varying amounts of p-hydroxybenzoic acid melt in the general range of 800.degree. to 900.degree. F. Obviously, the melting point of these polymers is far too high and the thermal stability is insufficient to permit these polymers to be used in conventional melt-processing equipment," PA0 "was to incorporate a substituent on some of the aromatic rings of the polyester, preferably on the diol ring. For example, it is well known that use of chloro, methyl or ethyl hydroquinone lowers the melting point of the polyester. Although this approach can be used to lower the melting point, typically the mechanical properties are also substantially reduced." PA0 "For example, liquid crystal copolyesters have been prepared from the following fairly rigid molecular species: p-hydroxybenzoic acid, hydroquinone and isophthalic acid. However, when ratios of the monomers are selected to provide tractable polymers, the glass transition temperature is generally low and the high temperature mechanical properties are reduced."
It is further stated (column 2, lines 33-40) that a solution to the above problem
The patent goes on to propose the use of phenyl hydroquinone (an expensive comonomer) as the best way whereby the melting point can be reduced to obtain tractable resins, without adversely affecting the mechanical properties. As indicated earlier, polyesters forming oriented melts were made from a variety of substituted hydroquinones. See, for example, Lee et al., U.S. Pat. No. 4,600,765; Hutchings et al., U.S. Pat. Nos. 4,614,790 and 4,614,791; and Funakoshi et al., U.S. Pat. No. 4,447,593. Readily processible polyesters made from p-hydroxybenzoic acid, isophthalic and optionally terephthalic acid, hydroquinone and 3,4'- and/or 4,4'-biphenol, 3,4'- and/or 4,4'-dihydroxydiphenyl ether, 3,4'- and/or 4,4'-dihydroxydiphenyl sulfide are the subject of Dicke et al., U.S. Pat. No. 4,603,190. It should be recognized that once again an expensive monomer is necessary to obtain tractable melts. Similar situations are encountered in a host of other U.S. and foreign patents. See, for example: Portugall et al., European Patent Appln. No. EP-257,558; Hisgen et al., European Patent Appln. No. EP-257,598; Hisgen et al., German Patent Appln. No. DE-3,629,208; Hisgen et al., German Patent Appln. No. DE-3,629,210; and Okamoto et al., World Patent Application No. WO-88/00,955.
As pointed out above, tractable materials result at high isophthalic acid levels but the products typically display undesirably low glass transition temperatures. Deex, U.S. Pat. No. 4,377,681 states (column 1, lines 31-38):
Attempts to increase the Tg of these products have been made. Thus, Deex, U.S. Pat. No. 4,377,681, claims copolyesters prepared from p-hydroxybenzoic acid, isophthalic acid, hydroquinone and 2,2-bis(4-hydroxyphenyl)propane. The preferred compositions contain from about 20 to about 35 mole percent of p-hydroxybenzoic acid units, and from about 5 to about 12 mole percent of 2,2-bis-(4-hydroxyphenyl)propane (bisphenol-A) based on the total diphenol components. Glass transition temperatures of about 175.degree. to about 190.degree. C. were observed in these polymers. These values represent an improvement when compared to the Tg's of the polyesters mentioned supra. However, they must be considered low as they lead to heat distortion temperatures (HDT's) which are, at best, of the order of about 120.degree. to 140.degree. C; moreover, the introduction of bisphenol-A lowers the degree of crystallinity and the rate of crystallization which we believe, as will be discussed infra, leads to lower HDT's. In addition, mold shrinkage of these copolymers is unsatisfactorily high.
Park U.S. Pat. No. 4,816,552 is directed to polymers prepared by heating a mixture of a 4-acetoxybenzoic acid derivative, a hydroquinone diacetate derivative, a dicarboxylic acid, and a 4-acetoxy-(4-acetoxyphenyl)benzoate derivative in two steps to remove generated acetic acid and form the polymer.
This patent is directed specifically to block copolymers with a specific sequence of recurring units. It reportedly has improved spinning processability. Our present invention is produced in a manner well known in the art to produce a random copolymer. As shall be shown infra, our invention is produced by the acidolysis method as disclosed in U.S. Pat. No. 3,637,595 which Park acknowledges produces a "partial block," i.e., a random copolymer. Park further states that polymers made by that method have no practical use due to high melting point, high viscosity, and inferior processability (U.S. Pat. No. 4,816,552 col. 2, l. 15-32).
The dilemna facing those who have attempted the development of tractable, high HDT, hydroquinone/benzene dicarboxylic acid/p-hydroxybenzoic acid copolyesters is perhaps best illustrated by Example 1 of Cottis et al., U.S. Pat. No. 3,975,487. A polyester having excellent properties, based upon a 1:3 molar ratio of isophthalic:terephthalic acids, p-hydroxybenzoic acid, and 4,4'-biphenol was prepared. When this example was repeated using hydroquinone in place of biphenol and all isophthalic acid as the dicarboxylic acid, a polymer having poorer properties (i.e., a lower flexural strength and a lower modulus) was obtained (column 10, lines 60-63).
Copolyesters based on p-hydroxybenzoic acid (PHBA), hydroquinone (HQ), isophthalic (IA) and terephthalic (TA) acids are disclosed in Cottis et al., U.S. Pat. No. 3,637,595. Cottis shows one example in which a resin incorporating all of the four monomers is described (Example 10). The polymer was formed from 1.0 mole of PHBA, 0.5 moles of IA, 0.5 moles of TA, and 1.0 moles of HQ. It was poorly characterized; weight loss in air, at 752.degree. F. was apparently high, indicating thermal stability problems. Our own characterization of the resin produced by the preferred process utilized in this invention (see Experimental, Comparative Example 2 also designated as X) shows that it possessed a low HDT (214.degree. C.)).
Thus, the elusive goal of developing a low cost hydroquinone-based crystalline polymer which when filled with 30 percent by weight of glass fibers (1) has an HDT of at least 240.degree. C. and preferably 280.degree. C. and higher, (2) is melt-processible below the decomposition temperature of about 415.degree. C., (3) has a melting point in the range of 340.degree. to 400.degree. C., a crystallization temperature of 300.degree. to 340.degree. C. and a crystallization rate of at least 1,000 counts per minute, has not been achieved in the prior art. In fact, based on the prior art, it appears unlikely that polymers having these properties can be produced, particularly polymers consisting essentially of units (I), (II), (III), and (IV). Furthermore, even though the overall combination of properties of the neat unblended polymers is outstanding, they may form molded parts that show undesirable blistering (i.e., raised areas).
It has now been discovered that the addition of a first polyester polymer (a) comprising recurring moieties of dihydroxyarylene comprising a hydroquinone, a nonvicinal benzene dicarboxylate (preferably terephthalic acid and mixtures of terephthalic acid and isophthalic acid) and p-oxybenzoate to a second polyester polymer (b) comprising recurring moieties of naphthalene based monomers and/or diphenol, nonvicinal benzene dicarboxylate and p-oxybenzoate, wherein said polymers and the moieties making up the polymers are present in specified proportions, yields allows in which the tendency to blister is substantially reduced or eliminated and also having increased strength and good fabricability.
The preferred mole ratios of monomers of the base polyester are depicted in the triangular diagrams and are described infra. The polymers melt in the range of from about 300.degree. to about 420.degree. C., preferably 340.degree.-400.degree. C. Of particular interest are the polyesters falling into area A of FIG. 1.
With some known exceptions, mixtures of polymeric materials are generally immiscible. That is, they consist of domains of chemically distinct phases. Usually, one component forms a continuous phase, while the other component forms roughly spherical domains as inclusions. Under some circumstances, bi-continuous structures are also obtainable. Mixing two arbitrarily chosen polymers usually results in inferior materials having no utility, since in the absence of adhesion between phases, the dispersed phase merely weakens the continuous phase. Some polymeric products, such as the wholly aromatic polyesters, exhibit an ordered structure in at least some regions of the polymer. This order can exist in one, two or three dimensions. The incorporation into blends of polymers exhibiting an ordered structure leads to an increased tendency of the blends to separate into phases. This is believed to be due to the fact that the order found in certain regions of the resin causes a fairly sharp boundary between the domains of the molecules of the component polymers. Hence, blends including such polymers would be expected to exhibit a significant reduction in properties.
It should be noted, however, that many useful blends whose morphology and phase interaction are favorable, are known.
Cottis, U.S. Pat. No. 4,563,508, is directed to the improvement of molding compounds based on wholly aromatic polyesters by the addition of a minor amount of a flow modifier. The flow modifier crystallizes poorly and improves the flow of the highly crystallized base polymer it is added to. The flow modifier does not enhance the end properties of the blend composition. It is to be noted that the addition of the flow modifier decreases the HDT of the composition and does not increase the strength.
Takayanagi et al., U.S. Pat. No. 4,228,218, discloses a polymer composition comprising 20 percent or less, based upon the total weight of polymeric material, of a first rigid polymer with the balance being a second polymer composed substantially of flexible molecular chains. The first polymeric material is dispersed in the second polymeric material in a microscopic region of 1 .mu.m or less. It is believed that wholly aromatic polyesters would be characterized by those skilled in the art as rigid molecules within the context of the above cited patent. The patent does not disclose blends of two or more polymers having rigid chains with improved blister resistance as does the present invention.
Blends of polymers exhibiting orientation in the melt with other polymers were investigated. Mixtures of liquid crystalline polyesters with poly(alkylene terephthalates), polycarbonates and polyarylates were described in Cincotta et al., U.S. Pat. Nos. 4,408,022 and 4,451,611; Froix, U.S. Pat. Nos. 4,489,190 and 4,460,735; and in Kiss, European Patent Application No. 169,947. Improved mechanical properties were found with these materials. The addition of a particular liquid crystal polymer to poly(butylene terephthalate) or other thermoplastic polymers was described as a method to obtain compositions with enhanced resistance to melt dripping during burning (see Kim et al., U.S. Pat. No. 4,439,578). In several instances, e.g., in alloys of liquid crystalline polyesters with an aromatic sulfone polymer (Froix et al., U.S. Pat. No. 4,460,736) with an aromatic poly(ester amide) (Kiss, U.S. Pat. No. 4,567,227), and with poly(arylene sulfides) (Froix, U.S. Pat. No. 4,276,397) improved mechanical characteristics and improved processibility (lower viscosity) of the non-anisotropic resin were noted. Better properties were also obtained by blending two particular liquid crystalline polyesters (see, for example, Froix, U.S. Pat. No. 4,267,289).
Liquid crystalline materials, including polyesters, were used to decrease the viscosity and improve the processibility of a number of other resins, including fluorinated polyolefins (see Bailey et al., U.S. Pat. No. 4,417,020; Cogswell et al., U.S. Pat. Nos. 4,429,078 and 4,438,236; and George et al., U.S. Pat. No. 4,650,836).
In one instance (Bailey et al., U.S. Pat. No. 4,508,891), it was claimed that the addition of an isotropic resin to an anisotropic resin leads to a decrease of anisotropy in the corresponding molded articles.
The fracture-surface morphology of thermotropic 6-hydroxy-2-naphthoic acid-p-hydroxybenzoic acid copolymer blends with nylon 6, poly(butylene terephthalate), and polycarbonate prepared by screw injection molding, was studied by Beery et al. J. Mater. Sci. Lett. 1988, 7(10), pp. 1071-3. The morphology was found to be strongly dependent on the flow history and on the composition of the subject mixtures.
A commonly assigned patent application entitled "Extrusion-Grade Compositions Comprising Mixtures of Wholly Aromatic Polyesters," Ser. No. 060,038, filed on June 9, 1987, in the names of Field et al., now U.S. Pat. No. 4,851,480, hereby incorporated by reference, describes alloys of a first polyester comprising recurring moieties of 4,4'-biphenol, terephthalate, and p-oxybenzoate; with a second polyester comprising the same recurring moieties, but wherein the proportion of the p-oxybenzoate units is higher than in the first polyester. The application discloses that while each individual polyester is difficult to extrude into acceptable products, their alloys provide good extrusion grade compositions. Molding compositions comprised of the above first and second polyester, filler, and optionally a polymeric flow modifier are claimed in commonly assigned U.S. Patent Application entitled "Molding Compositions Comprising Mixtures of Wholly Aromatic Polyesters and Fillers," Ser. No. 060,114 filed on June 9, 1987 in the name of J. J. Duska, hereby incorporated by reference.
Thus, it is known from the prior art that it is possible to alloy two polyesters, wherein said polyesters are based on identical monomers but differ in the relative proportion of the monomers, wherein each of said polyesters has unsatisfactory molding and extrusion characteristics and obtain good molding and extrusion grade compositions.
No reference is known which is directed to the improvement of surface properties (i.e., blistering) by blending two polymers having orientation in the melt. A feature of the instant invention that is totally unexpected and highly remarkable is the fact that compatible blends showing good mechanical and surface properties are achieved by alloying two crystalline wholly aromatic copolyesters prepared from monomers having totally different structures, e.g., phenylene versus naphthalene or biphenylene. The enhanced blister resistance is particularly astonishing. Indeed, while the blister phenomenon is not fully understood, it has been attributed to non-homogeneity in random copolymers. In any event, as indicated earlier, alloys of materials having ordered structures would be expected to have reduced properties. Hence, the instant discovery was highly surprising and totally unexpected.
It is the general object of the present invention to provide novel, inexpensive, melt-processible hydroquinone poly(iso-terephthalates) containing residues of p-hydroxybenzoic acid polymers which form a highly tractable oriented melt phase, and which are capable of melt extrusion to form quality, high performance fibers, films, three-dimensional molded articles, etc. The polymers display high HDT's which are essential in certain high heat applications. They have good flexural strength and produce good fibers. When blended with other polymers they produce inexpensive high heat polymers capable of high HDT's and still produce an attractive blister-resistant surface.
It is a further object of the present invention to provide novel, melt-processible hydroquinone poly(iso-terephthalates) containing residues of p-hydroxybenzoic acid, polymers which form a melt phase below 400.degree. C. in the substantial absence of polymer degradation, unlike many other polymers which include relatively high concentrations of the 4-oxybenzoyl moiety.
These and other objects, as well as the scope, nature and utilization of the invention will be apparent to those skilled in the art from the following detailed description.